Optical tomographic mapping of the human somatosensory cortexJan Mehnert1,2, Susanne Holtze1,2 , Stefan P. Koch1 , Christina Habermehl1, Christoph Schmitz1,3, Arno Villringer1,2,4, Jens Steinbrink1

1Berlin NeuroImaging Center, Charité University Medicine Berlin, Germany

2Max-Planck Institute for Human Cognitive and Brain Sciences Leipzig, Germany3NIRx Medizintechnik GmbH, Berlin, Germany

4 Berlin School of Mind and Brain, Humboldt-University Berlin, Germany

DISCUSSION

• The results show that optically based high-density imaging is feasible to localize and

discern somatopic activations to vibrotactile stimulation of different fingers.

• As expected by homuncular organisation of the somatosensory cortex the hemodynamic

response to vibrotactile stimulation of the thumb was localized more laterally compared

to the 5th finger. This result is in good agreement with fMRI studies that have

investigated the human somatosensory system [2,3,4] and proves that functional optical

techniques can yield high-resolution maps of functional cortical anatomy.

NEXT STEPS

• statistical validation of the activation spots for the different conditions

• alignment of functional tomographical data and structural MR

REFERENCES [1] Zeff, B.W., White, B.R., Dehghani, H., Schlaggar, B.L. & Culver, J.P. (2007), ‘Retinotopic mapping of adult human visual cortex with high-density

diffuse optical tomography’, Proc Natl Acad Sci, vol. 104, no. 29, pp. 12169-74.

[2] Kurth, R., Villringer, K., Mackert, B.M., Schwiemann, J., Braun, J., Curio, G., Villringer, A. & Wolf, K.J. (1998), ‘fMRI assessment of somatotopy in

human Brodmann area 3b by electrical finger stimulation’, Neuroreport, vol. 9, no. 2, pp. 207-12.

[3] Kurth, R., Villringer, K., Curio, G., Wolf, K.J., Krause, T., Repenthin, J., Schwiemann, J., Deuchert, M. & Villringer A. (2000), ‘fMRI shows multiple

somatotopic digit representations in human primary somatosensory cortex’, Neuroreport, vol. 11, no. 7, pp. 1487-91.

[4] Ruben, J., Schwiemann, J., Deuchert, M., Meyer, R., Krause, T., Curio, G., Villringer, K., Kurth, R. & Villringer A. (2001), ‘Somatotopic organization

of human secondary somatosensory cortex’, Cereb Cortex, vol. 11, no. 5, pp. 463-73.

CONTACT

mehnert@cbs.mpg.de

RESULTS

• strongest functional changes to finger-tapping and vibrotactile stimulation at deeper

tomographical slices, [Fig. 7]

• arbitrary thresholding of T-values reveal distinct ‘centers of mass‘ for finger tapping

and vibrotactile stimulation in 6 of 8 subjects, [Fig. 8 & Fig 9]

[i] finger tapping leads to a more anterior located activation compared to

vibrotactile stimulation and

[ii] activation for thumb is more superior located compared to 5th finger

BACKGROUNDNear-infrared Spectroscopy (NIRS) is a versatile functional imaging tool of great flexibility

• Besides of advantages (interference-free, low cost, portability) a major shortcoming of

this methodology is the low spatial resolution

• Improvement of spatial resolution can be obtained by increased probe-density and usage

of the multi-distance approach

• Zeff et al. [1] showed retinotopic activations in the human visual cortex to eccentric and

rotating stimuli by high-density optical tomography.

• Here we investigate whether high-resolution optical topography allows to demonstrate

homuncular somatotopic representation in the human somatosensory cortex.

IMAGING SETUP

• rectangular array of optical probes (30 fibres) were attached over right somatosensory

region (C4 according to 10-20 system), [Fig. 2]

• Because each probe is source and detector, the setup provides 900 measuring channels and

allows optical tomography (multi distance approach)

• NIRS imaging system: DYNOT 232 (NIRx Medizintechnik GmbH, Berlin, Germany;

wavelengths: 760 nm & 830 nm, sampling frequency: 2.44 Hz

METHODSSTIMULATION PROCEDURE

• thumb and the 5th finger of the left hand were

stimulated pseudo-randomly with PC-controlled

electrical toothbrushes integrated in a glove (8

subjects; 5 male), [Fig. 1]

• 4 s vibrotacitle stimulation + 12-16 s baseline

period (on & offsets indicated by tone)

• 20 repetitions for each condition

DATA ANALYSIS

• band pass filtered time series (0.03 Hz to 0.4 Hz) were converted to tomographical

hemodynamic changes (HbR and HbO) using NAVI (NIRx N.Y.; FEM model), [Fig. 3]

• General Linear Model for finger tapping (1 predictor) and somatosensory stimulation (2

predictors) on reduced tomographical data, [Fig. 4]

• optical volumes were sliced tangentially to the surface (~45°), [Fig. 5]; this was also done

for MR-volume, [Fig. 6]

[Fig.4] [Fig. 5] [Fig. 6]

ai

s

l m

p

5th finger

thumb

finger tapping

vibro

2

6

..48

slices outside volume slices outside volume1 ..2 1 ..2

.. 45 .. 45

superficial

depth

subject C subject D

[Fig. 7] T-values for HbR parameter arbitrary thresholded for each condition from 2

subjects; panels depict tomographical slices from superficial to deeper slices

4s

s2

s1

m

MS1

S2

finger tappingvibrotactile stimulation

• subjects also performed randomized self-paced finger tapping to allow localization of sensori-

motor cortex (left hand, 4s )

• Task performance was guided by the acoustic sound of the toothbrushes, thus minimizing

differences between the somatosensory and the motor task

ANATOMICAL MR:

• All subjects had an anatomical MR-scan with fiducial markers

[Fig. 3]

[Fig. 1]

i

s

p a

i

s

p a

[Fig. 9] HbR changes for finger tapping and

vibrotactile stimulation overlayed on

anatomic MR for one subject. Strongest

activity to finger tapping is located at the

precentral gyrus. Vibrotactile stimulation

leads to activation at the postcentral gyrus.

Subject 1 Subject 2 Subject 3 Subject 4subj. A subj. B subj. C subj. D subj. E

[Fig. 8] medium depth (5 subjects): T-values for HbR (arbitrary threshold)